Use of principal factor analysis to generate a corn silage fermentative quality index to rank well- or poorly preserved forages

BACKGROUND: To investigate corn silage fermentative quality, a principal factor analysis was carried out on a database consisting of 196 corn silages sampled from the core, lateral and apical parts of silo feed-out face and characterised by 36 variables. Eleven principal factor components (PCs) were retained and interpreted. Two PCs were related to chemical and digestibility variables; four PCs were characterised by end-products associated with clostridia, heterolactic, homolactic or aerobic fermentations; two PCs were associated withmycotoxins produced by Penicillium roqueforti or by Aspergillus fumigatus and Fusariumspp., while three PCs explained ensiling procedures adopted to store corn silages. RESULTS: Lower (P<0.05) yeast ormould counts and greater (P<0.05) aerobic stabilityweremeasured in core than in peripheral samples. Excluding PCs related to ensiling procedures, other PCswere able to predictmicrobiological counts, aerobic stability or biogenic amine content as verifiedby multiple linear regression analysis. Based onthese results, several corn silage quality index calculations were computed by using a summative equation approach in which different PCs as well as diverse relative weights multiplying each PCs were combined. To compute definitive index calculation, only PCs explaining clostridia, heterolactic and homolactic fermentations were used with relative weights of 30%, 50% and 20%. CONCLUSIONS: The new proposed fermentative quality index was highly correlated to parameters related to corn silage fermentative quality, such as microbiological counts, aerobic stability or biogenic amines and it properly discriminated welland poorly preserved forages.